Unraveling the thermodynamics of molecular recognition
Smolinski, Michael P
MetadataShow full item record
Specific molecular recognition is a ubiquitous phenomenon essential to the operation of biological processes. The process by which two entities recognize each other and bind in a specific phenomenon is a very complex process. Computer algorithms have been written in an attempt to describe and predict molecular recognition with the goal of aiding rational drug design efforts. The success of these algorithms has been inconsistent. It is the aim of this work is to generate consistent sets of data that go beyond just training sets and address some of the poorly understood phenomenon associated with molecular recognition, namely enthalpy-entropy compensation and cooperativity. The serine proteases thrombin and trypsin were selected as the receptor. A number of ligands have been carefully designed to examine cooperativity. The binding of these ligands has been characterized by X-ray crystallography and isothermal titration calorimetry. The series of ligands were designed to maintain a consistent binding mode as the amount of hydrophobic surface area and salt-bridging ability varies. Although much of the calorimetric and crystallographic experiments are still being conducted, some very interesting data has come from a few of the ligands. In the process of binding to thrombin, two ligands; N-cyclopentylglycine-proline-para-amidinobenzamide and N-cyclohexylglycine-proline-para-amidinobenzamide, show a classic case of enthalpy-entropy compensation. They bind with the same binding affinity, however the driving forces and crystallography of the complexes show significant differences. The cyclopentyl analog has a distinct binding mode and has an even balance of favorable enthalpy and entropy whereas the cyclohexyl analog shows disorder in its binding mode and is more entropically driven. More refinement of the above data and further calorimetric experiments are being conducted to help explain the reason for these differences. Many other compounds have been designed and synthesized to examine the roots and mechanisms of cooperativity. The biological testing of these compounds is still in progress. It is believed that the major impendent to the success of computer algorithms predicting the affinity and specificity of molecular recognition is a poor understanding of some of the subtle global phenomenon associated with molecular recognition. Properly designed and executed experiments aimed at examining these phenomenons may greatly improve our understanding which may lead to significant advances in some of nature's greatest mysteries. Beyond drug receptor binding the additional fields affected by this work may include protein and nucleic acid folding, protein:protein interactions and how to modulate them with small molecules, as well as a better understanding of more complex, dynamic systems like GPCRs that have already been shown to be very useful therapeutic targets.